Peer39’s proprietary technology is designed to deliver actionable semantic data at massive scale on the fly. Our process is comprised of 3 main steps:
1. Capture - Our Grabber extracts only the relevant content from any web page
- Peer39’s innovative algorithm scans each page for relevant sections and paragraphs based on text, style (fonts, size, color), and location on page
- Relevant content on the page is extracted and sent to the classifier, which results in highly relevant classifications
2. Classify - Our semantic content classification is driven by Natural Language Processing
- Using NLP, our technology analyzes content from across the web, identifying similarities and learning how to associate strings of words and phrases with the right category
3. Deliver - Peer39 offers three easy ways to implement and integrate our solution
- API - HTTP call based for server to server integration.
- Batch – An asynchronous way of integration, with the ability to process batch files of traffic and upload the results back to the client.